IoT as a Service

To ensure secure access to the data held in internet of things, many lightweight authentication schemes have been developed using approaches such as symmetric cryptography or hashing operations. Although these schemes achieve forward key secrecy and user anonymity, de-synchronization is a major problem in these protocols. As such, many other schemes have been presented to address this pertinent security challenge. However, some of these schemes are still susceptible to smart card loss attacks among others. In this paper, stochastic security ephemeral generation protocol for 5G enabled internet of things is presented. It is demonstrated to offer mutual authentication and session key agreement. It is also robust against packet replays, eavesdropping and man-in-the-middle attacks. In terms of performance, it has the lowest computation and communication overheads.

The ubiquitous nature of the Internet of Things (IoT) constitutes a set of stringent quality of service (QoS) requirements from the underlying 5G network. To address the issues, this paper proposes an enhancement for the channel access period (CAP) of the widely used IEEE 802.15.3C (millimeter wave (mmW)) standard using a priority mechanism that fulfills the requirement of prioritized channel access for the IoT based applications. According to the hybrid medium access control (MAC) protocol of the IEEE 802.15.3C, to reserve time-division multiple access (TDMA) based slot in channel time allocation period (CTAP), a node will first send a channel time allocation (CTA) request to piconet controller (PNC) by using carrier-sense multiple access with collision avoidance (CSMA/CA) mechanism in the contention-access period (CAP). After successful delivery of CTA’s request, PNC will reserve a CTA for a specific node. However, there is no guarantee that a node will get a channel in the contention process in the existing standard. Hence, the existing CAP mechanism could demonstrate a bottleneck for a data sending device in terms of less delay and high throughput. To solve this issue, we first design a numerical model of CAP using the IEEE 802.15.3C standard’s specification, and then we propose a priority-based mechanism with three priority classes: high priority (HP), medium priority (MP), and low priority (LP) with each class having different contention windows (CW) range that makes the value of backoff period shorter. To evaluate the performance of the proposed mechanism, modifications are applied to the proposed numerical model. The performance comparison is conducted among prioritized classes devices in terms of transmission delay, channel access delay, and throughput. The conducted evaluations include two types of data rates i.e., 1.5 Gbps and 3 Gbps. The proposed scheme shows promising results for a node that requires high priority in an IoT environment.

Channel estimation is crucial for a millimeter wave MIMO system. Due to the existence of massive antenna elements, the overhead to perform channel estimation with traditional methods would be huge, which will degrade the throughput severely. Thanks to the sparsity of channel model on millimeter wave band, most existing literature make use of this feature to compress the number of signaling based on the technique of compressive sensing. In this paper, by making use of the fact that the angle of arrival (AoA) and angle of departure (AoD) vary much slower than the channel coefficients, we go one step forward on saving the number of signaling for channel measurement. Specifically, with a consideration of channel sparsity feature, we design a set of methods to detect the variation of AoA and AoD in time, which includes the case of appearance of new path and disappearance of existing path, through sequential analysis approach. Moreover, to enhance the performance of our proposed method, procoder and combiner are designed respectively to generate beam on anticipated directions, through semi-definite programming method. With the above operations, we only need to measure channel coefficients when the AoA and AoD are not detected to change, which does not require much signaling. Through this way, the overhead for channel measurement is further saved compared with the methods based on compressive sensing.

The Internet of things (IoT) technologies have been developing since their inception. Consequently, the number of connected devices increases yearly. The development of IoT devices has to be set, taking into consideration parameters such as security, data rate and energy. In this paper, we carried out a comprehensive review on the main concern, which is the energy efficacy of IoT devices. We will target four research areas to make the searching process interesting and easier for researchers. The four research areas are related to clustering and routing in WSNs, smart grid, fog computing and MQTT & CoAP application protocols.

The concept and applications of the Internet of Things or IoT are well-known to those dealing with the technicalities and complexities of IoT. However, for most users, the understanding seems to be limited to the benefits and usability of the devices. In particular, grasping the privacy, security and other relevant issues, especially social issues, remains out of reach for most users. This paper addresses the problem of privacy, security and other relevant issues from users’ perspective and suggests three areas needing greater attention in resolving the issues. First, this paper highlights social issues and emphasizes the role of business leaders in dealing with the issues surrounding IoT devices. This paper argues that the onus and obligation lie with the business leaders as social architects to perform their duty of care in a socially responsible manner. Second, IoT is simply an IS product in which people and their views are one of the key elements for achieving the common goal, in this case, of networking of things and people. Ignoring the role of end-users as a critical part of IoT does not help achieve the common purpose. Lastly, given the transnational nature of the issue, governments worldwide are essential stakeholders and hence need to have a proactive and positive approach in the fight against the use of IoT for cybercrimes.

The Internet of Things (IoT) is a relatively new concept as it connects things (or objects) that do not have high computational power. The IoT helps these things see, listen, and take action by interoperating with minimal human intervention to make people’s lives easier. However, these systems are vulnerable to attacks and security threats that could potentially undermine consumer confidence in them. For this reason, it is critical to understand the characteristics of IoT security and their requirements before starting to discuss how to protect them. In this scope, the present work reviews the importance of security in IoT applications, factors that restrict the use of traditional security methods to protect IoTs, and the basic requirements necessary to judge them as secure environments.

The fundamental goal and commitment of this article is the exploitation of our perception-based intelligent management method. An examination with 39 elective methodologies was performed, exhibiting the upsides of our methodology as far as interpret able and precise fuzzy principle-based DSGC strength forecast then revealing the chain of importance of DSGC-framework’s characteristic criticalness. Shrewd networks are strong, self-recuperating systems that authorized bidirectional circulation of vitality and data inside the utility framework. Therefore, prosumer supervision involves growing attentiveness between scholars in current years. At that point, this evaluation process of nearby market interest is tackled by deep reinforcement learning and deep Qlearning techniques with experience replay system. This idea discovers the safety of upcoming energy frameworks touching near to coordinating extra parts of sustainable power source elements. Particularly, we can manage cold-start clients with less social connections. Later on, we will distinguish further data from informal community to viably tackle client cold-start issues. Moreover, we will investigate the effect of complex data on client utilization conduct to assemble a stable recommender system. The subsequent part presents logical examinations of Internet of Things (IoT)applications in the power business. For the logical examination relevant investigation, brilliant local area meter information-driven and autonomous models are made to figure the likely kilowatt (kW) limit decline from DR. At long last, I bring up open inquiries to empower further exploration.

The Internet of things (IoT) concept can be generally described as the ability of machines to communicate via the Internet to perform tasks. In addition to the communication between devices, humans can remotely control IoT devices via controllers such as smartphones. The main aim of introducing IoT technologies is to make our lives easier and more convenient. Due to the massive increase in both IoT devices and research on enhancing the security and the speed of these devices, there is a strong demand to work in parallel to promote IoT networks’ energy efficiency to make IoT systems scalable. This paper outlines the causes of energy inefficiency in IoT systems and proposes some key tools to prolong the network lifetime of these devices.

Although many studies are conducted for ICT systems for educational organisations, there is a lack of understanding of ICT systems’ usage for school administration. Infrastructure IT support Communication Experience and Training (IiCE) is a proposed framework based on the ICT that subsumes the IoT intending to improve the primary school administration tasks. This study is conducted on Saudi public primary schools. This study aims to investigate the current state of ICT systems used in primary schools’ administration to provide a framework for the ICT systems usage. We carried out an electronic survey and analysed more than 500 responses. We include a detailed analysis reflecting different school members roles, including teachers, school principals, administrative assistants and parents. The findings of this study highlight the limitations in the existing ICT systems for school administration and provide a holistic understanding of influencing factors. Decision-makers for Saudi educational organisations can use the findings with the proposed framework to better understand the current situation of ICT systems and provide better solutions.

The study analyzes the impact of heterogeneous institutional shareholding on corporate social responsibility (CSR) performance and corporate internal control (IC), and explores the mediating effect of internal control between heterogeneous institutional shareholding and CSR performance. The results show that Pressure-resistant institutional shareholding significantly improves CSR fulfillment degree and the effectiveness of internal control, while Pressure-sensitive institutional shareholding has no significant promoting effect on CSR fulfillment and IC effectiveness. Effective IC has a significant mediating effect between Pressure-sensitive institutional shareholding and CSR fulfillment. Finally, it is suggested that the structure of institutional investors should be optimized, and the shareholding ratio of Securities investment funds, qualified foreign institutional investors (QFII) and Social security funds should be enhanced to improve CSR performance and IC effectiveness. And the mediating effect of IC between institutional ownership and CSR fulfillment should be promoted, to strengthen the consciousness of enterprises to take the initiative to enhance CSR performance.

Service-level agreement (SLA) is the commitment between consumers and providers. Service providers commit the consumers to assure the quality, availability and responsibility. SLA renegotiation is negotiation between consumers and providers to improve the quality of service. Cloud computing service provided by providers can be guaranteed with SLA. Therefore, SLA renegotiation is vital to service and profit of both providers and consumers. This article presents some detailed analysis of previous papers and the comparison of components of theirs. In this paper, an improved framework is proposed to solve the existing problems of traditional SLA renegotiation process. In the end, limitation and future work are concluded.

Stock exchanges are economic entities facilitating various trading assets like monetary values, activities, valuable metals, etc., among stockbroker participants. Prediction of Stock market rates and observing the behaviour of daily closing rates is a crucial task for many businesses and investment authorities. This acts as a precaution to know the suitable period for stakeholders to invest. Deep Learning, in this regard, is considered to perform forecasting tasks efficiently with better accuracy. For this purpose, our study performs forecasting of Australian Stock Market daily closing rates based on Deep Learning approaches of LSTM and GRU from January 4 2000, to January 17 2017. This work predicts the closing rates for the next 216 days. A comparative analysis of prediction accuracy between Deep Learning methods like Long Short-Term Memory (LSTM) along with Gated Recurrent Unit (GRU) is performed. Results reveal that the deep learning model LSTM performs better than the other approach based on the results obtained. Performance of the models is measured using metrics such as RMSE and R² scores, where LSTM achieved a comparatively less RMSE value of 0.072 and the largest R² score of 0.855.

Nowadays, IoT monitoring systems are ubiquitous. These systems utilized sensors to measure the temperature indoors or outdoor. These sensors can be temporarily unavailable for several reasons, such as power outages. Thus, the server that collects the temperatures should find an alternative for predicting the temperature during the downtime of temperature sensors. In this context, there are several machine learning models for predicting temperature. This work is motivated to study the performance gap of predicting outdoor and indoor temperatures. In the proposed study, we utilized a deep learning recurrent neural network called Gated Recurrent Units (GRUs) and four machine learning models, namely, random forest (RF), decision trees (DT), support vector machines (SVM), and linear regression (LR) for predicting the temperature during the downtimes of the temperature sensors. Then, we evaluated the proposed models on a realistic dataset. The results show that predicting the indoor temperature is more predictable than the outdoor temperature. Moreover, the results revealed that the SVM model was the most accurate model for this task.

AutoML tools provide an automation service for data scientists and software engineers to save time from data preprocessing and modeling building. Existing AutoML tools usually require users to have data science knowledge and programming skills to use the services, however, most non-expert and business users do not have such skills to use these AutoML tools. In addition, many AutoML tools require a special infrastructure or cloud provider. In this paper, we introduce BrewAI: a commercial-grade tool that provides an easy-to-use AutoML service for business users. The paper describes how the use of service-oriented computing design principles gives BrewAI flexibility, scalability and performance at a reasonable cost. The paper also describes a case study that shows how BrewAI enables business users to outperform more than three-quarters of Kaggle competitors in an NLP classification task.

Internet of things (IoT) based monitoring systems became commonplace. These systems are built upon a large number of devices and sensors. The data collection task of a large number of sensors and devices in an IoT system includes a massive number of data communications. The more the number of devices, the critical is the network bottleneck. In this context, the dual prediction scheme was proposed as a solution for mitigating the large size of communication volumes. The dual prediction scheme consists of a model for predicting future measurements based on historical data. This model is duplicated on both sides, the edge side (i.e., sensor) and the data collection device (i.e., cluster head). The literature includes several works which proposed many dual prediction schemes based on several techniques such as filters and moving average. The literature does not include utilizing the ensemble learning models. This motivates this work to investigate the gradient boosting regression model’s performance compared to the existing solutions. The proposed and state-of-the-art models are evaluated on a realistic dataset. The obtained results show that the proposed model outperforms the existing dual prediction schemes in terms of communication reduction.

With the tremendous increase in product alternatives these days, many businesses rely heavily on recommender systems to limit the number of options they display to their customers on the front end. Many companies use the collaborative filtering algorithm and provide suggestions based on other consumers’ choices, like the active user. However, this approach faces a cold start problem and is not suitable for one-time transactions. Thus, this research aims to create a recommender system that uses online customer reviews in the IoT framework to match the attributes of a product important to the shopper. The algorithm makes recommendations by first identifying the product’s features essential to a customer. It then performs aspect-based sentiment analysis to identify those features in customer reviews and give them a sentiment score. Each customer review is weighted based on its creditably. As the impact of the recommender systems varies with the product type, an experimental study will be carried out to study the effect of the proposed algorithm differs with hedonic and utilitarian products.

Wind energy is becoming one of the most important suppliers of renewable energy but due to its reliance on weather conditions it is highly inconsistent and its integration into electricity grids is a challenge. In this research we present a comparative analysis of the performance of several prominent data mining techniques in prediction of wind energy generation. Data from the Big Data Challenge Bremen 2018 was used for short term forecasting. Of basic models, a decision tree produced the best performing model. It performed marginally better than SGD, OLS, LASSO and Bayesian ridge regression. Whereas, SVM, nearest neighbor and Gaussian NB performed very poorly. A further analysis using ensemble methods was performed where a Gradient Boosting was the best model. Further improvements of the IoT model are performed and limitations of this are discussed in detail.